1
The Impacts of Information Technology on Total Factor Productivity:
A Look at Externalities and Innovations
Yen-Chun Choua, Howard Hao-Chun Chuang
a, and Benjamin B. M. Shao
b,*
aDepartment of Management Information Systems
College of Commerce
National Chengchi University
Taipei 11605, Taiwan, ROC
bDepartment of Information Systems
W. P. Carey School of Business
Arizona State University
Tempe, AZ 85287-4606, USA
Abstract
The impacts of information technology (IT) on total factor productivity (TFP) are assessed through an
integrative framework of IT-induced externalities and IT-leveraged innovations. Based on network
externalities and endogenous growth theory, our study aims to reconcile the seeming discrepancy between
the recent observed evidence and the prediction by neoclassical growth theory. We argue that
computerization has reshaped the competitive landscape into a network economy with IT-induced
externalities that benefit not only IT purchasers but also other stakeholders. Moreover, IT is a platform
technology that can leverage innovations to enhance the technological level of production process.
Consequently, these two factors of IT-induced externalities and IT-leveraged innovations exert positive
impacts on TFP, suggesting IT plays a more pivotal role than input consumption and accumulation that
neoclassical growth theory assumes for IT. We use panel data from 30 Organization of Economic
Cooperation and Development (OECD) countries over the period of 2000 to 2009 to empirically test
hypotheses on this IT-TFP link. Implications are drawn from our findings for future research to measure
IT’s contributions at the macro level more accurately, and policymakers are urged to cultivate IT’s
positive impacts on TFP to help sustain long-term economic growth.
Keywords: impacts of information technology; total factor productivity; network externalities;
endogenous growth theory; innovations; production economics.
Revised June 2014
_________________________ *Corresponding author. Tel.: + 1 480 727 6790; fax: +1 480 727 0881.
E-mail address: [email protected] (B.B.M. Shao).
2
1. Introduction
Information technology (IT) has recently been identified as a key driver for economic growth
(Jorgenson, 2001; Stiroh, 2002). To explain the contributions made by IT to output production,
researchers have leveraged neoclassical growth theory and treated IT as one type of input similar to
capital and labor used in the production process that contributes to value added at the firm level
(Brynjolfsson and Hitt, 1995, 1996) and gross domestic product (GDP) at the country level (Dewan and
Kraemer, 2000). Moreover, since IT capital goods have become much less expensive over the years, its
cost advantage has led to factor substitution for other inputs (Dewan and Min, 1997; Lin and Shao, 2006a;
Chwelos et al., 2010), IT capital deepening (Oliner and Sichel, 2000; Jorgenson et al., 2008), and better
performance of the IT industry to meet increasing demands (Chou et al., 2012). From the perspective of
neoclassical growth theory, IT increases economic outputs through input consumption and accumulation.
Distinct from single factor productivity like labor productivity (Demeter et al., 2011), total factor
productivity (TFP) is a performance metric that measures the “residual” portion of output produced
beyond mere input usage (Hulten, 2001; Comin, 2008). TFP is driven by non-input factors such as
technological progress that are not tied to explicit input usage (Stiroh, 2002). For example, when a
producer combines resources in an innovative way to produce more output or create new features of value,
this technological progress is reflected in TFP. Given the law of diminishing returns of inputs, TFP must
continuously improve in order to sustain an economy’s long-term growth through means like
technological progress that can offset diminishing returns of input usage. However, since neoclassical
growth theory conjectures that IT’s contribution to output only occurs through input usage, IT has been
regarded as unrelated to TFP by the related studies based on this framework (Brynjolfsson and Hitt, 2003).
The recent TFP acceleration of the U.S. economy, however, was found to be dominated by the
industries that use IT creatively and innovatively (Jorgenson et al., 2006). This observation leads to the
proposition that IT’s effects on output may extend beyond input usage (Bosworth and Triplett, 2007). IT,
for instance, may exert a direct influence on TFP through IT-induced externalities based on network
effect as computerization has reshaped the competitive landscape into a network economy that provides
3
value not only to IT purchasers but also to other stakeholders. An example is the IT investments made by
buyers and suppliers of business-to-business (B2B) e-commerce systems that can be beneficial to all
parties involved. Cheng and Nault (2007) and Han et al. (2011) examine the impact of inter-industry IT
spillover on technology level of production processes and find that IT intensity and competitiveness
moderate the effects of internal IT investments. Potential network effects resulting from communication
systems and software use are other instances of IT-induced externalities (Brynjolfson and Kemerer, 1996;
Oren and Smith, 1981; Shy, 2001).
In addition to IT-induced externalities, IT-leveraged innovations may also improve TFP in such
forms as IT-facilitated new business processes and integrated supply chains (Kim and Narasimhan, 2002;
Kim et al., 2011; Brynjolfsson and Hitt, 2000, 2003). IT services and IT-based product innovations may
have a similar effect (Chou and Shao, 2014; Davenport et al., 2008). Park et al. (2007) have extended the
idea of IT-leveraged innovations to knowledge transfer across countries through international trade of IT
goods. In this light, IT as a platform technology can complement innovations to generate synergies and
enhance the technological level of production process (Shapiro and Varian, 1999).
Our study probes two central research questions. First, does IT have impacts on TFP? Second, in
what forms or through what channels do such impacts of IT manifest themselves? Based on network
externalities (Katz and Shapiro 1985, 1986) and endogenous growth theory (Romer, 1986, 1990), our
study reconciles the seeming discrepancy between the recent observed evidence of TFP and the prediction
by the conventional neoclassical growth theory. We empirically explore the relationship between IT and
TFP to show that IT enhances TFP through IT-induced externalities and IT-leveraged innovations, thus
extending beyond the traditional role of IT as input consumption and accumulation that is deemed
orthogonal to TFP by neoclassical growth theory. To answer the specific research questions, we analyze a
panel dataset collected for 30 Organization of Economic Cooperation and Development (OECD)
countries over the period of 2000 to 2009 and report our empirical findings.
Our paper contributes to the literature in four ways. First, on the theoretical front, we assess the
economic value of IT beyond what neoclassical growth theory implies and explore the relationship
4
between IT and TFP through the lens of IT-induced externalities and IT-leveraged innovations. Second,
on the empirical front, our more recent data enables an assessment of IT-induced externalities that might
not be prominent in the pre- or early Internet eras for which prior studies had produced mixed or
inconclusive results (OECD, 2000; Stiroh, 2002). Third, while most existing studies on the topic focus on
the U.S. (Stiroh, 2002; Han et al., 2011), our investigation into 30 OECD countries (including the U.S.)
provides a broader perspective and addresses the shortage of cross-county studies on the association
between macro characteristics and the economic value of IT (Lin, 2009). Finally, our measure of IT
capital includes not only computer hardware but also software and telecommunication equipment. This
operationalization of IT capital provides a more comprehensive measurement for examining IT-induced
externalities and IT-leveraged innovations whereas prior studies mostly focus on computer hardware
(Cheng and Nault, 2007; Han et al., 2011).
Dedrick et al. (2003) note that the impacts of IT on outputs in terms of marginal product and
labor productivity enhancement through capital deepening have been extensively studied and well
understood; however, IT’s spillover effects and interplays with other complementary assets are still
unsubstantiated and this represents an opportunity for more inquiry. In particular, it is “difficult to
precisely interpret the allocation of productivity improvements to capital deepening and to MFP”
(Dedrick et al., 2003, p.16). Our study endeavors to answer this call for more research on these issues by
assessing IT’s impacts on TFP and providing empirical evidences that so far have been elusive in the
literature.
Our study relates to the productivity paradox coined by Solow (1987) that questioned the real
contributions of IT to output production at the country level. Since it was raised, researchers had long
debated about whether IT improves output production and renders its process more productive, and they
examined the issue from different perspectives and analyzed data from a variety of sources. Reviews of
this paradox can be found in Berndt and Hulten (2007), Brynjolfsson (1993), Diewert and Fox (2001),
Lucas (1999), and Triplett (1999), among others. After positive and significant evidences were reported at
the firm level (Brynjolfsson and Hitt, 1996; Hitt and Brynjolfsson, 1996) and the country level (Dewan
5
and Kraemer, 2000; Jorgenson, 2001; Oliner and Sichel, 2000), the productivity paradox as originally
formulated is now considered resolved (Dedrick et al., 2003). It is noted that the productivity paradox is
cast in the framework of neoclassical growth theory and hence performance metrics being used by prior
studies include marginal product, output elasticity, and substitution elasticity of IT, not TFP. Therefore,
our IT-TFP study builds upon the prior literature and aims to shed new light on the issue.
The rest of the paper is organized as follows. Section 2 articulates the theoretical arguments for
the IT-TFP link and formulates two hypotheses for IT-induced externalities and IT-leveraged innovations.
Section 3 presents our research model that makes the IT-TFP link explicit and empirically testable.
Section 4 describes the data collection and variable definitions. Section 5 presents estimation results, and
Section 6 discusses the findings and implications. Finally, Section 7 concludes the paper with limitations
and topics for future research.
2. Theoretical Framework and Hypothesis Development
2.1. IT-induced Externalities
In economics, externality is a benefit or cost experienced by one party due to another party’s
actions, without compensating payment between the parties. A network is a system made up of
interconnected users who wish to communicate with one another. Network effect exits when a network’s
value to any user depends on the number of other users on the network, and positive network effect means
the value of joining a network increases with more users. Network externalities are externalities due to
network effects (Katz and Shapiro, 1985). In other words, network externalities occur when the value of
owning and using a network product is directly linked to the installed base of the product (Hajji et al.,
2012). Products that exhibit network effects are called network goods. Network effects are a powerful
source of value and oftentimes called the “demand-side economies of scale” as they influence the revenue
side of a provider’s profit equation by increasing users’ willingness-to-pay for its products or services.
They are in contrast to the more traditional “supply-side economies of scale” that improve profit margins
of a provider by reducing unit costs through means like fixed cost leverage, division of labor, and
6
experience effect (Shapiro and Varian, 1999).
Many IT products and services are network goods that enjoy network effects and hence create not
only direct value for their purchasers but also network externalities for other stakeholders involved.
Examples of IT-induced network externalities include emails, mobile phones, video games, social
networks, enterprise resource planning (ERP) systems, and proprietary software, to name a few. In other
words, network externalities can be observed in the domains using these IT products and services (Katz
and Shapiro, 1986; Rohlfs, 2001). For instance, Sena (2004) finds evidence for knowledge spillover from
high-tech firms to non-high-tech firms in manufacturing by enhancing their total factor productivity. Hajji
et al. (2012) note that network externalities of ERP systems lead to continuous updating of marketing and
pricing strategies in the launching and early phases of market penetration for new ERP systems.
In our study, we argue IT contributes to TFP through such IT-induced externalities derived from
the use of IT goods that have network effects. This channel of IT-induced externalities is elucidated by
the new economy paradigm that envisages the outcomes of “more creates even more” and “co-creation of
value in a network” as the IT investment made by one party improves the productivity of another party
(Kelly, 1999). The new economy paradigm regards IT as the means to engender network externalities that
constitute a different channel – other than input consumption and accumulation – for IT to contribute to
economic growth (Stiroh, 1999). In this light, IT-induced externalities go beyond factor contributions and
impact TFP directly.
One representative example of IT-induced externalities is the connectivity observed in an
integrated supply chain wherein returns to IT investments made by one party are rendered more valuable
by IT investments of other trading partners (Kim and Narasimhan, 2002; Melville and Ramirez, 2008). In
this setting, a supplier’s capability to accurately forecast demand for raw materials through the supply
chain relies on the demand data provided by its buyer’s IT systems. Similarly, a buyer utilizes fulfillment
data from its supplier’s IT systems to make speedy and accurate order replenishment. Such connectivity
makes IT investments beneficial not only to the owner of investments but also to its trading partners in
the integrated supply chain. Accordingly, IT capabilities and information sharing are both found to have
7
positive effects on logistics integration (Prajogo and Olhager, 2012), and IT use along with customer and
supplier integration is considered an important organizational competency (Jitpaiboon et al, 2013).
Another manifestation of IT-induced externalities is the network effects observed in many IT
goods and services. Gandal (1994) and Brynjofsson and Kemerer (1996) find that the value of a software
to a user increases due to network externalities from the community of users. Also, the benefits of a party
joining an electronic market derive not only from its IT equipment purchase and sign-up fee but also from
others’ participation (Bakos, 1991). We argue that these “non-pecuniary externalities” from IT usage are
not reflected in factor productivity of IT capital goods but in their aggregated contributions to TFP
(Schreyer, 2000).
The Internet is another prominent IT system that generates significant network externalities. In
the late 1990s, the Internet started being used as a legitimate and efficient platform to conduct business
transactions. Since then, externalities from the Internet usage have become even more pronounced over
time. Compared with its predecessor Electronic Data Interchange (EDI), the Internet incurs much lower
coordination costs. Before the Internet, early forms of B2B e-commerce were mainly conducted through
EDI. Due to its prohibitively high set-up costs and lock-in effects, EDI was limited to supply chain
coordination between large businesses and their first-tier suppliers (OECD, 2000). The Internet, however,
has greatly facilitated B2B e-commerce by lowering development costs and switching costs. B2B e-
commerce thus became feasible for firms of all sizes once the Internet was accepted as a legitimate
platform for performing various business functions such as procurements, distributions, and customer
relationship management. Moreover, the Internet also significantly reduces transaction costs for firms to
participate in electronic markets. As a consequence, IT-induced externalities are expected to be more
prevalent at the present than in the pre- and early Internet eras for which studies could only find
inconclusive or mixed results (Stiroh, 2002).
Overall, IT-induced externalities are based on the idea that social returns to IT usage exceed
private returns of ownership (Delong and Summers, 1991), thus leading to spillover benefits of network
effects for other stakeholders and to an aggregated enhancement of TFP at the country level. Based on the
8
above arguments, we formulate our first hypothesis on IT-induced externalities:
H1: Countries with a high level of IT capital have high TFP.
2.2. IT-leveraged Innovations
According to neoclassical growth theory, IT is considered unrelated to TFP, the part of output
beyond input utilization (Stiroh, 2002). In this view, TFP is driven by technological progress fueled by
innovations that are spontaneous, unpredictable, and independent of firms’ decisions and government’s
policies. The argument for this seeming disconnect is that the trajectory of technological advance is
difficult to predict as it does not follow straightforward evolutionary path. Although neoclassical growth
theory recognizes the importance of technological progress to economic growth, it considers innovative
advancement to be unrelated to firms’ and government’s actions (Grossman and Helpman, 1994). Thus,
one limitation of the theory is that it places technological progress of innovations, a deciding factor of
economic development and competitiveness, on the periphery (Miozzo and Walsh, 2006).
This neoclassical view applied to IT as a factor of production has been predominant in the
productivity literature. Using this paradigm, most previous studies treat IT as a capital input unrelated to
TFP and they assess the value of IT by analyzing its marginal product contribution to outputs (e.g., Barua
et al, 1995; Brynjolfsson and Hitt, 1996; Dewan and Kraemer, 2000). Grounded in neoclassical growth
theory, these studies also note that rapid decline in quality-adjusted prices of IT capital goods leads to
substitution for other input factors and hence to IT capital deepening (Stiroh, 2002). In sum, IT
contributes to output production and factor productivity through input consumption, substitution, and
capital deepening.
In reality, however, firms purposively invest in technologies to adopt new innovations when they
see opportunities to profit. Contrary to neoclassical growth theory, endogenous growth theory contends
that technological development reflects firms’ and government’s conscientious and diligent efforts
(Romer, 1986, 1990). Technological level thus is influenced by these decision makers’ collective
endeavors to continually introduce innovations to production systems that can result in either new
9
products and services, new rules of doing business like mass customization and devolved manufacturing,
or new capabilities to collaborate with other parties (Albadvi et al., 2007; Miozzo and Walsh, 2006;
Jorgensen, 2009).
In endogenous growth theory, innovations are a crucial driver for technological development of
production process. Since IT is a general-purpose platform technology for leveraging innovations
(Bresnahan and Trajtenberg, 1995; Helpman, 1998; Varian et al., 2004), it can enhance technological
development of production process and lead to an increase in TFP. Therefore, in addition to its impact on
TFP through IT-induced externalities, IT can also play a facilitator role in helping leverage innovations
for synergies. Like steam engines and electricity in the past, IT is considered by many to be the latest
platform technology because it satisfies the criteria of rapidly falling costs, plentiful supply, and
numerous applications to products and processes (Miozzo and Walsh, 2006).
This leverage of innovations by IT has been alluded to and suggested by the literature.
Responding to the claim made by McKinsey (2002) about the missing correlation between productivity
and IT investments, Martinsons and Martinsons (2002) urge researchers to explore IT value beyond
efficiency (Lin and Shao, 2000, 2006b; Lin and Chiang, 2011) and factor contributions (Brynjolfsson and
Hitt, 1995) and to focus on the effects of IT on innovations. McAfee and Brynjolffson (2008) also
promote this innovation channel of IT value by arguing that “IT serves two distinct roles—as a catalyst
for innovative ideas and as an engine for delivering them.” They further note that it is the innovation role
of IT that affects performance differentials among competitors. Varian et al. (2004) find that as more
transactions become computer mediated, the costs of activity monitoring and coordination become much
lower, and this IT-based capability complements innovative products and processes, reshaping the
competitive landscape for industries like retailing, entertainment, and transportation. IT-leveraged
innovations can lead to flexibility, quick response, variety, custom fit, high performance, and speedy
delivery, all of which are reflected in TFP.
Aggregated at the macro level, Freeman and Perez (1988) suggest that IT is a general-purpose
platform technology that can leverage new innovations in products, services and processes, and bring
10
about significant impacts on the economy. As a consequence, IT facilitates the use of creative ideas
embedded in knowledge-based assets and promotes innovative development of other products and
processes (Calestous and Lee, 2005; Joshi et al., 2010). That is, IT can be a complement to innovations
due to its capabilities and applications to product development and process improvement. Accordingly,
we present our second hypothesis on IT-leveraged innovations:
H2: Countries with a high level of IT-innovation complementarity have high TFP.
3. Research Model
To empirically test the IT–TFP link, we propose a production function model based on the growth
accounting framework in economics (Oliner and Sichel, 2000). We assume a general production function
f that relates an economy’s output (Q) to inputs of labor (L), IT capital (KC), non-IT capital (KN), and
technology parameter (A):
𝑄 = 𝑓(𝐿, 𝐾𝐶 , 𝐾𝑁, 𝐴) (1)
Under the assumptions of competitive markets, full input utilization, and constant returns to scale,
the contribution of each input to output is equal to the share of input cost to total costs (Schreyer, 2000).
By assuming a Cobb-Douglas functional form, we express the production function f in logarithms for
estimation as follows (lowercase letters for inputs denote logarithms and, for brevity but without loss of
clarity, country and time subscripts are suppressed):
𝑞 = 𝑠𝑙𝑙 + 𝑠𝑐 𝑘𝑐 + 𝑠𝑛 𝑘𝑛 + 𝑎 (2)
where parameter s represents each input’s factor share as indicated above.
In the presence of IT-induced externalities, there exists a discrepancy between a private investor’s
rate of return and the rate of return for society as a whole. In other words, IT generates benefits above and
beyond those reflected in its measured factor share. We modify Equation (2) to include IT-induced
externalities by a term θ (Romer, 1986; Lucas, 1988; Schreyer, 2000):
𝑞 = 𝑠𝑙𝑙 + (𝑠𝑐 + 𝜃) 𝑘𝑐 + 𝑠𝑛 𝑘𝑛 + 𝑎 (3)
IT-induced externalities θ is similar in spirit to knowledge spillovers studied by Romer (1986)
11
and spillovers from human capital looked at by Lucas (1988). In practice, it is difficult to observe θ
directly. What is normally observed is the factor share 𝑠𝑐. By definition, TFP is the difference between
output and input utilization. Hence, from Equation (3), we can derive TFP as follows:
𝑡𝑓𝑝 = 𝑞 − 𝑠𝑙𝑙 − 𝑠𝑐 𝑘𝑐 − 𝑠𝑛 𝑘𝑛 = 𝜃 𝑘𝑐 + 𝑎 (4)
In Equation (4), technology parameter a captures a country’s technological advancement.
According to endogenous growth theory, we model the technology parameter a as a linear function of a
country’s innovation efforts in the form of patents and other related factors:
𝑎 = 𝐷𝑖 + 𝛽1𝑃𝑎𝑡𝑒𝑛𝑡 + 𝛽2 (𝑘𝑐 × 𝑃𝑎𝑡𝑒𝑛𝑡) + 𝛽3𝐻𝐷𝐼 + 𝛽4𝐹𝐷𝐼 + 𝜀 (5)
In Equation (5), Di is a dummy for country-specific fixed effect that captures unobserved
heterogeneity. The variable Patent is the number of patents that reflects a country’s innovation efforts
since the patent is generally considered an intermediate outcome of innovations for the commercialization
of new products and services (Nerkar and Shane, 2007). The interaction term kc×Patent represents the
synergy that results from the complementary leverage of patents by IT (i.e., IT-leveraged innovations).
HDI stands for human development index that indicates a country’s level of human capital, which is
critical to technological innovation capabilities (Furman et al., 2002). FDI denotes foreign direct
investment that reflects openness of a country for receiving external investments in building new
production facilities. Openness benefits a country’s technological development because it allows a
country, through foreign investments in facility establishment and production means introduction, to have
access to new knowledge and competence from other countries, especially from those that are
technologically more advanced (Glass and Saggi, 1998). Finally, ε is the error term that captures both
random error and technical inefficiency (Lin and Shao, 2000).
By incorporating Equation (5) into Equation (4), we reach the complete IT-TFP model:
𝑡𝑓𝑝 = 𝐷𝑖 + 𝜃 𝑘𝑐 + 𝛽1𝑃𝑎𝑡𝑒𝑛𝑡 + 𝛽2(𝑘𝑐 × 𝑃𝑎𝑡𝑒𝑛𝑡) + 𝛽3𝐻𝐷𝐼 + 𝛽4𝐹𝐷𝐼 + 𝜀 (6)
Equation (6) indicates that TFP is influenced by IT-induced externalities θ kc (i.e., Hypothesis H1)
and IT-leveraged innovations kc×Patent (i.e., Hypthesis H2) along with other relevant factors (i.e., Patent,
HDI, and FDI). To provide a visual overview, Figure 1 presents our IT-TFP research framework with the
12
two hypotheses specified.
Figure 1. The IT-TFP Research Framework
4. Data Description
We collected a panel dataset for 30 OECD countries from 2000 to 2009, including Australia,
Austria, Belgium, Canada, Czech Republic, Demark, Finland, France, Germany, Greece, Hungary,
Ireland, Israel, Italy, Japan, Korea, Mexico, New Zealand, Netherlands, Norway, Poland, Portugal,
Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, the United Kingdom, and the United States. The
period of the dataset is specifically chosen for the new millennium because by 2000, network effects of IT
products and services were conceivably manifested when the Internet had matured as an e-commerce
platform. The list of countries is selected based on data availability of their IT investments. We acquired
archival data on TFP, IT capital, and other variables from several data sources. Variables measured in
monetary terms are converted into 1995 international dollars by purchasing power parity (PPP).
The measure of TFP is derived from the EU Total Economy Database. The estimation of TFP
growth in the database is carried out using the growth accounting framework. According to the EU Total
Economy Database, TFP growth captures the changes in output not caused by changes in inputs. TFP thus
H1(+)
IT Capital
HDI
TFP
Patent
FDI
H2(+)
13
reflects the effect of technological change, efficiency improvements, and the inability to measure the
contribution of all other possible inputs. It is calculated as Solow residual by subtracting the sum of two-
period average compensation share weighted input growth rates from the output growth rate. Log
differences of level are used for growth rates, and hence TFP growth rates are essentially Tornqvist
indexes. In calculating Solow residual, the Cobb-Douglas function is used most often because of its
analytical tractability but in theory any production function with constant returns to scale that satisfies the
Inada conditions will do (Romer, 2000).
Using 1995 as the base year, we construct TFP on the basis of growth data from the database in
the following way: TFPt+1= TFPt×TFPGrowtht. Each country’s TFP is further transformed to the level
relative to the U.S. for cross-country comparisons. A scale multiplier is determined by a country’s labor
productivity relative to the U.S.’s from the OECD Productivity Database.
The data on IT capital stock at the country level are known to be difficult to collect because only
a limited number of national statistical offices in OECD countries have published capital stock data on a
regular basis. In addition, the comparison across countries is not straightforward due to the different
aggregate measures used across volume indices, asset types, and industries (Schreyer and Webb, 2006).
Under this constraint, we approximated IT capital stock values from a series of IT spending. We acquired
IT investment data for each country from World Information and Technology and Service Alliance
(WITSA, 2010). This data provides a consistent basis for comparison of IT capital, including computer
hardware, software, and communication equipment across countries.
To estimate IT capital stock from IT investment data, we used the method suggested by the
OECD’s Productivity Measurement Manual (OECD, 2008). The method enables the estimation of an
initial capital stock in a given year, working forward from the earliest investment flows that one is able to
acquire, to derive the accumulated capital stocks in subsequent years. The initial capital stock at t = 0 for
country i is defined as TotalCapitali0 = InvFlowi0/(gi+δ), where InvFlowi0 is the initial capital investment
flow in the benchmark year for country i; the growth rate gi of investment capital flow is approximated by
the average GDP growth rate over the period of 1990 to 1999; and the depreciation rate δ for IT goods is
14
obtained from the EU KLEMS database. Once we obtained the initial capital stock TotalCapitali0 for each
country, we computed the subsequent IT capital stock series using the perpetual inventory method
(OECD, 2008): TotalCapitalit = TotalCapitali(t-1)×(1–δ) + InvFlowi(t-1). We adopted this perpetual
inventory method to construct three sets of capitals for computer hardware, software, and communication
equipment. IT capital is then aggregated from these three data series and used as one independent variable
for our model estimation.
Regarding the other variables in Equation (5), Patent represents the number of patents granted
from the U.S. Patent and Trade office (USPTO) and the European Patent Office (EPO). Patent is
commonly used for measuring a country’s innovation capacity (Mansfield, 1985) and it is highly
correlated with a country’s R&D investments (Furman et al., 2002; Park et al., 2007). We also collected
data on HDI from the Human Development Report Office and obtained data on FDI from the
Euromonitor database. Table 1 presents descriptive statistics of the variables.
Table 1. Descriptive Statistics of Variables
Variable Obs. Measurement Unit Mean Std Dev Min Max
TFP 293 Index 0.800 0.247 0.277 1.377
IT Capital 293 $ Million 147,000 325,000 2,840 1,790,000
Patent 293 Number of Patents 120,875 352,849 32 2,246,346
HDI 293 Index 0.843 0.057 0.629 0.937
FDI 293 $ Million 230,000 424,000 1,470 2,720,000
5. Estimation Results
To estimate Equation (6) using panel data, it is customary to specify the regression model as yit =
βxit + ci + uit where ci captures the individual heterogeneity and uit represents the idiosyncratic error. The
difference between fixed effect (FE) modeling and random effect (RE) modeling lies in the correlation
between explanatory variables xit and heterogeneity ci where FE allows corr(xit, ci) ≠ 0 but RE assumes
corr(xit, ci) = 0 (Wooldridge, 2001). Although RE modeling is more efficient, the Hausman test (Cameron
and Trivedi, 2009) rejected the null hypothesis that no systematic difference exists in FE and RE
coefficients. Hence, we chose the FE model over the RE model and allowed country-specific
15
heterogeneity (ci) to correlate with IT capital as well as other regressors. Following the approach of two-
way FE modeling (Cameron and Trivedi, 2009), we also added year dummies to control for time effect.
Since we have data over 10 years from 2000 to 2009, we conducted the Wooldridge’s test for
autocorrelation in pane data models (Wooldridge, 2001) and found a significant first-order autocorrelation
in error terms εit. We also performed a likelihood ratio test for panel-level heteroskedasticity (Greene,
2000) and rejected the null hypothesis of homoskedasticity in error terms. Taken together, we adopted a
procedure of robust clustered standard errors (Hoechle, 2007) to ensure estimation consistency in the
presence of autocorrelation and heteroskedasticity. Furthermore, we were aware of the potential existence
of contemporaneous correlations between countries. The Pesaran’s CD test (2004) revealed no evidence
of cross-sectional dependence in our data. We estimated Equation (6) using the XTREG command with
the cluster option in Stata 13 (StataCorp, 2013).
Table 2. Coefficient Estimates of the Models
Model I Model II Model III Model IV Model V
Intercept -0.607
(0.608)
-0.639
(0.626)
-0.986*
(0.569)
-1.040
(0.679)
-1.085
(0.684)
IT 0.056
**
(0.025)
0.057**
(0.025)
0.065***
(0.020)
0.064***
(0.021)
0.064***
(0.020)
Patent 0.0001
***
(0.000)
0.0001***
(0.000)
0.0001***
(0.000)
0.0001***
(0.000)
IT × Patent 0.002
*
(0.001)
0.002*
(0.001)
0.002*
(0.001)
HDI 0.092
(0.593)
0.130
(0.603)
FDI 0.000
(0.000)
F-statistic
(df1, df2)
8.66
(10, 29)
12.46
(11, 29)
11.81
(12, 29)
10.95
(13, 29)
11.78
(13, 29)
p-value 0.00 0.00 0.00 0.00 0.00
R2 0.34 0.36 0.41 0.41 0.42
Country 30 30 30 30 30
Observations 293 293 293 293 293
Note: Standard errors in parentheses; *p < 0.10,
**p < 0.05,
***p < 0.01.
Table 2 presents the estimation results where all the five models are statistically significant
according to the F-test. Including only IT capital and FE terms, Model I shows a significant positive
association between IT and TFP, a finding that is contrary to the orthogonality assumed by neoclassical
16
growth theory. Models II to V introduce four more variables (i.e., Patent, IT×Patent, HDI, and FDI)
stepwise to reflect a country’s innovation efforts as suggested by endogenous growth theory. As can be
seen, the coefficient estimates for IT capital remain significantly positive throughout and we have
acquired evidence for Hypothesis H1 that IT-induced externalities exist as IT capital has a direct positive
impact on TFP.
Next we find the coefficient estimates for IT×Patent in Models III-V are all significantly positive,
providing evidence for Hypothesis H2 that IT-leveraged innovations have a favorable impact on TFP. For
the other variables, Patent not surprisingly also has a positive direct effect on TFP. On the other hand,
neither HDI nor FDI is found to have any significant impact on TFP.
The presentation of Models I-V in Table 2 provides us with more results as consistent evidences
to support our two hypotheses on IT-induced externalities and IT-leveraged innovations. In other words,
instead of relying on one particular model, having these different models with each introducing an
additional variable shows more evidences if the results on the two main variables of interest IT and
IT×Patent are consistent across the models. Such use of different models to offer reliable results is often
employed by researchers in reporting their findings (e.g., Chang and Gurbaxani, 2013; Chari et al., 2008;
Ghose et al., 2006).
To ensure that the identified IT-TFP link is robust, we further assess the potential endogeneity
between IT capital and TFP. That is, when externalities captured by TFP are high, countries may be more
inclined to make investments in IT. The endogeneity may violate the assumption of corr(ITit, uit) = 0 and
cause estimation bias. Using ITCapitali,t-1 as an instrument for ITCapitalit (Brynjolfsson and Hitt, 1996),
we checked the endogeneity issue by performing the Davidson-MacKinnon test on the null hypothesis
that FE estimates are consistent (Cameron and Trivedi, 2005) as well as the C-test that is robust to
heteroskedasticity (Baum et al., 2007). Both tests revealed no significant endogeneity. Furthermore,
consistent findings are derived from instrumental variable estimation with two-stage least square
(Wooldridge, 2001) and two-step generalized method of moments that accommodates autocorrelation and
heteroskedasiticty (Baum et al., 2007). These robustness checks on potential endogeneity issues further
17
validate our findings in Table 2.
Table 3 summarizes our results in terms of hypotheses formulated, theories used, outcomes, and
the main findings. Overall, our analysis shows that IT capital is positively associated with TFP. The
estimated effects are fairly stable across all the regression models. Our country-level results provide
empirical evidence for the hypothesis that IT-induced externalities contribute to TFP. In addition, we also
find evidence that IT-leveraged innovations increase TFP as the interaction between IT capital and patents
creates synergies and aids economic growth.
Table 3. Results Summary
Hypothesis Theory Outcome Finding
H1: Countries with a high
level of IT capital
have high TFP.
Network
Externalities Supported
IT-induced externalities
have a positive impact on
TFP.
H2: Countries with a high
level of IT-innovation
complementarity have
high TFP.
Endogenous
Growth Theory Supported
IT-leveraged innovations
have a positive impact on
TFP.
6. Discussions and Implications
In neoclassical growth theory, TFP is deemed unrelated to input usage. Accordingly, IT capital as
a production input contributes to factor productivity by its factor share but is considered orthogonal to
TFP. The new economy paradigm based on network effects, however, suggests that IT capital as a special
production input may go beyond factor share and contribute to TFP through IT-induced externalities.
Additionally, according to endogenous growth theory, IT may also play a role of complement to
innovations and provide synergies that add to TFP.
In formulating and testing the two hypotheses on IT-induced externalities and IT-leveraged
innovations, our study empirically confirms the IT-TFP link and demonstrates the economic value of IT
beyond input consumption and accumulation. Although researchers have speculated about the possible
presence of externalities from IT usage, few studies have explored this phenomenon empirically
(McKinsey, 2002; Lin, 2009). Using archival data on 30 OECD countries from 2000 to 2009, our analysis
18
presents new evidence for the under-studied association between IT investments and productivity at the
macro level. This identified IT-TFP link is corroborated by our study as substantial and robust since our
models also consider other possible innovation factors (i.e., patents, HDI, and FDI) that may influence
TFP as suggested by endogenous growth theory.
Growth in economic outputs can be achieved either by increases in inputs or by improvements in
TFP. Relying exclusively on input increases, an economy will experience output increases only in
proportion to input consumptions in capital and labor. In the long run, the law of diminishing returns of
inputs would cause economic growth to slow down or even stop eventually (Solow, 1956). In this regard,
TFP is of paramount importance because it is the antidote to such adverse effects of diminishing returns
of input usage in order to sustain long-term economic growth. Along the same line, Hall and Jones (1999)
find major differences in income per capita between rich and poor countries to be associated with
differences in TFP. Based on the concept of network externalities, our empirical findings indicate that IT
enhances not only factor productivity but also TFP. Policymakers thus should adapt their decision-making
and recognize the spillover value of IT in TFP as an important factor for sustaining long-term economic
growth. In terms of measurement, conventional metrics focus on economic impacts of IT through IT
capital deepening and mostly ignore the contributions of IT to TFP. Our results show that efforts should
be made to measure, reflect and capture additional economic benefits of IT more accurately as in TFP and
innovations.
In terms of practical implications, we can see from our confirmation of the IT-TFP link that it is
imperative for a government to devote attention and resources to the generation, diffusion and use of IT
goods and services in an economy. IT represents a new technological paradigm shift that belongs to the
category of general-purpose technologies (Miozzo and Walsh, 2006). As such, IT has the potential to be
adopted and adapted in a wide range of sectors in a country, in ways that can dramatically improve
operations and products as well as the relationships between different sectors. In this regard, IT serves
both roles of competence enabler and capability enhancer (Perunović et al., 2012). As a competence
enabler, IT opens up new opportunities or markets, and as a capability enhancer, IT helps deliver better
19
solutions that are cheaper and faster. A government thus should seek regulation reforms and allocate
resources to develop IT competence and capability using various means, such as fostering skill
development through the education and training in science, technology, engineering, and math (STEM).
The leverage of IT to support innovations also suggests innovational complementarities where the
productivity of a downstream sector increases as a result of innovation in an upstream sector facilitated by
IT capabilities. These innovational complementarities underscores the importance of establishing a
national innovation system where IT can help connect and cluster institutions, policies, and practices that
together determine a country’s capacity to generate and apply innovations.
Moreover, since innovations can be either incremental or radical (Abernathy, 1978), IT should be
used to leverage innovations in two different ways as well. For incremental innovations that refine,
improve and exploit existing innovations to strengthen the dominance of incumbents, IT can be used to
facilitate certain sequential improvements sought by these incremental innovations. One good example of
using IT to introduce incremental innovations is Netflix, the company that has been dominating the
industry of Internet video streaming and DVD rental over the last decade with its recommendation system
Cinematch and dynamic queue, subscription and delivery methods. For radical innovations that offer
dramatic improvement in performance or cost and that result in transformation of existing markets or
creation of new ones, IT can play a pivotal role in enabling the fundamental changes in the design or
platform of the innovative products and processes. One illustrative case of continuously introducing IT-
enabled radical innovations is Google, the company that revolutionizes Internet search, online advertising,
cloud computing, and other phenomena with its many new products, services, software, and offerings.
On the front of network externalities, IT products and services that exhibit network effects yield
spillover benefits to all the parties involved in the transaction or connected in the system. Since many IT
goods and services exhibit positive network effects and have direct impacts on TFP, a government ought
to find ways to encourage their adoption and use and to facilitate their dissemination through the economy.
It can design and offer the right incentives to help industries and sectors adopt and use more IT goods and
services as well as best practices. It can also eliminate legislative and administrative hurdles that stymie
20
IT adoption and usage by, for example, assisting in establishing IT standards and protocols. Thus, IT
investments should be encouraged and effective incentive mechanisms should be designed and put in
place to induce network externalities from interconnection of IT equipment across firms and industries.
Our research framework hypothesize that IT exerts direct and interaction effects on TFP and our
empirical results support these hypotheses. In this light, IT is inherently different from other traditional
factors of production such as non-IT capital and labor that are considered unrelated to TFP by both
neoclassic growth theory and previous studies. It is noted that IT can play both roles of traditional capital
and knowledge capital (Dedrick et al., 2003). As a traditional capital, IT is used as a production
technology and subject to the same law of diminishing returns as non-IT capital and labor are. On the
other hand, as a knowledge capital, IT’s payoff through capital deepening and TFP growth suggest
increasing returns. It is in the second role of knowledge capital that IT functions differently from other
traditional inputs like non-IT capital and labor. Our results on the IT-TFP link complement the previous
findings that IT is a different as its use is associated with a technological shift that favors workers with
higher skills (i.e., skill-biased technical change) (Autor et al., 1998).
7. Conclusion
IT is believed to have a favorable impact on value creation by establishing linkages with trading
partners, customers, and suppliers (Soroor et al. 2009), resulting in better products, higher quality,
enhanced equipment utilization, reduced resource needs, and increased flexibility (Kim and Narasimhan
2002). Aggregated at the macro level, IT should have a positive effect on the production output of an
economy (Jorgenson, 2001). Our study examines the link between IT and TFP and corroborates the
existence of IT economic value beyond factor productivity.
Grounded in network externalities and endogenous growth theory, we propose an integrative IT-
TFP framework to explain the relationships between IT and TFP through IT-induced externalities and IT-
leveraged innovations. Our study empirically tests this integrative framework that links IT with TFP to
reconcile the seeming discrepancy between neoclassical theoretical arguments and recent TFP statistics.
21
We conclude that IT induces network externalities that benefit not just IT adopters but all stakeholders,
and technological progress reflects not only uncertainties faced in technology development but also
production agents’ conscious efforts to leverage innovations for knowledge dissemination and idea
applications (Malhotra et al., 2005).
Agarwal and Lucas (2005) suggest that demonstrating the value of IT investments is fundamental
to advancing our understanding of the subject. Kohli and Grover (2008) also identify the need for theory
and measurement development in IT value as one area for conducting research on the topic. Responding
to these calls, our IT-TFP study proposes an integrative framework of IT economic value that extends
beyond input consumption and accumulation. In essence, we empirically establish two unconventional
channels with externalities and innovations for IT to contribute to TFP. Our study represents one of the
first endeavors towards the direction for future research on IT economic value to go beyond factor shares
and partial productivity.
A substantial amount of empirical literature exists on IT capital deepening and economic outputs.
However, Dewan and Min (1997) point out that another interesting direction for future research is to
analyze time series data and try to separate the effects of factor substitution and technical change. With
the topic of IT factor contributions well studied, there is a need for more research on IT impacts at the
macro level. Our study considers the role of IT in enhancing TFP through IT-induced externalities and IT-
leveraged innovations. While a few prior studies have examined the impacts of IT on new products and
processes (Joshi et al., 2010), we take one step further and evaluate the complementarity of IT and
innovations for synergies in terms of TFP.
Moreover, we consider possible network externalities induced by IT in such forms as B2B e-
commerce, communication networks, and software usage, among others. The non-pecuniary externalities
from IT lead to direct contributions of IT to TFP. Our comprehensive measure of IT capital that includes
computer hardware, software, and telecommunication equipment provides a better instrument for
examining externalities from IT usage while most prior studies mainly focused on computer hardware.
We also study IT-induced externalities based on more recent data to explore such effects that might not be
22
observable or pronounced in the pre- or early Internet eras (Stiroh, 2002).
In conclusion, our research on IT value in TFP looks beyond neoclassical growth theory by
proposing and empirically testing an integrative framework that explains the relationships between IT and
TFP through the two channels of externalities and innovations. TFP represents the “residual” portion of
outputs above input usage, and our study indicates that TFP can be a competitive advantage that benefits
from IT-leveraged innovations. Therefore, our study offers a missing piece of the picture by linking IT’s
innovation effects to TFP. By answering the two research questions posed in the first section to fill the
gap in the literature, our study shows that IT exerts impacts on TFP and it does so through the two
channels of network externalities and innovation leverage.
In their critical review of the related literature on the topic of IT and economic performance,
Dedrick et al. (2013) propose a conceptual framework that emphasizes different perspectives at the three
levels of country, industry, and firm. While the inputs used are similar (e.g., IT capital, non-IT capital,
and labor), each level employs production processes subject to unique influential factors (e.g., capital
deepening, technical progress, and labor quality) that lead to different outcome measures (e.g., economic
growth, labor productivity, profitability, and consumer welfare). It is noted that the phenomena at each
level are interrelated and sometimes become the foundation for the phenomena at a higher level, but the
findings may not be directly portable, suggesting the need for researchers to conduct specific studies at
each level. For example, the finding that “IT investment improves aggregate productivity does not imply
that individual firms enjoy similar benefits… there may be significant social benefits from IT investments
that increase consumer welfare but are not captured by the firms making those investments” in the form
of profitability (Dedrick et al., 2003, p.7). At the same time, relatively fewer industry-level studies raise
the prevalence issue of IT payoff across industries (i.e., do IT investment payoffs occur across many
industries or are they concentrated to just a few?). Therefore, while our study is at the country level and
corroborates the impacts of IT on TFP, our findings need to be further substantiated for the industry, firm
or even individual worker level. It is likely, for instance, that actual IT usage which is easier to measure at
the firm and individual worker levels may be a key factor for TFP differences at those levels (Aral et al.,
23
2012; Devaraj and Kohli, 2003).
Like any research, our study has its limitations. First, in following the literature, we assess the
TFP impacts of IT-induced externalities and IT-leveraged innovations by controlling for other innovation-
related factors such as HDI and FDI. This list of controls is by no means exhaustive. We try the best to
econometrically alleviate this issue by incorporating year- and country-specific fixed effects into our
analysis. We encourage future studies to expand data collection and further validate the IT-TFP link.
Second, we recognize that country-level data are highly aggregated and inevitably subject to
measurement errors. For example, improvements in the quality of product may still be captured in the
output measure, even with the aid of hedonic price index (Hulten, 2001). In cases where quality
improvements of intermediate inputs are not perfectly measured, input contributions of downstream
industries can be manifested beyond factor shares, while they in fact benefit from better quality of
intermediate inputs and hence enjoy more output (Baily and Lawrence, 2001).
Finally, it is possible that the association between IT and TFP can be driven by other forces or
country characteristics in addition to IT-induced externalities and IT-leveraged innovations. Since it is not
feasible to include all potential variables in our models, this possible omission of relevant variables may
lead to endogeneity and estimation bias. We address this potential issue by conducting appropriate
statistical tests and by employing econometric techniques that can remedy potential biases in our model
estimation. Still, future research is encouraged to consider other relevant variables (either IT-related or
not) to identify other sources of TFP. One approach for doing so is suggested by Lin and Chiang (2011).
It will be interesting to follow the same approach and adopt similar estimation models to investigate the
interrelationships between IT and specific contextual factors of countries.
Acknowledgements
The authors appreciate the constructive comments and insightful suggestions provided by the anonymous
reviewers and the editor-in-chief, which have greatly improved the lucidity and quality of the paper. The
usual disclaimer applies.
24
References
Abernathy, W.J., 1978. The Productivity Dilemma: Roadblock to Innovation in the Automobile Industry.
Johns Hopkins University Press, Baltimore, MD.
Agarwal, R. and Lucas, H.C.. 2005. The information systems identity crisis: focusing on high-visibility
and high-impact research. MIS Quarterly, 29 (3), 381-398.
Albadvi, A., Keramati, A., and Razmi, J., 2007. Assessing the impact of information technology on firm
performance considering the role of intervening variables: organizational infrastructures and
business processes reengineering. International Journal of Production Research, 45 (12), 2697–
2734.
Aral, S., Brynjolfsson, E., and Van Alstyne, M., 2012. Information, technology, and information worker
productivity. Information Systems Research, 23 (3-2), 849-867.
Autor, D., Katz, L., and Krueger, A., 1998. Computing inequality: have computers changed the labor
market? Quarterly Journal of Economics, 113 (4), 1169-1213.
Baily, M.N. and Lawrence, R.Z., 2001. Do we have a new e-conomy? American Economic Review. 91
(2), 308-312.
Bakos, Y., 1991. A strategic analysis of electronic marketplaces. MIS Quarterly, 15 (3), 295-310.
Barua, A., Kriebel, C.H., and Mukhopadhyay, T., 1995. Information technologies and business value: An
analytic and empirical investigation. Information Systems Research. 6 (1), 3-23.
Baum, C.F., Schaffer, M.E., and Stillman, S., 2007. Enhanced routines for instrumental variables:
generalized method of moments estimation and testing. Stata Journal, 7 (4), 465-506.
Berndt, E. and Hulten, C. (eds), 2007. Hard-to-Measure Goods and Services: Essays in Honor of Zvi
Griliches. University of Chicago Press, Chicago.
Bosworth, B.P. and Triplett, J.E., 2007. The 21st century productivity expansion still in services?
International Productivity Monitor, 14 (Spring), 3-19.
Bresnahan, T. and Trajtenberg, M., 1995. General purpose technologies: engines of growth? Journal of
Econometrics, 65 (1), 83-108.
Brynjolfsson, E., 1993. The productivity paradox of information technology. Communications of the
ACM, 36 (12), 66-77.
Brynjolfsson, E. and Hitt, L.M., 1995. Information technology as a factor of production: the role of
differences among firms. Economics of Innovation and New Technology, 3 (4), 183-200.
Brynjolfsson, E. and Hitt, L.M., 1996. Paradox lost? firm-level evidence on returns to information
systems spending. Management Science, 42 (4), 541-558.
Brynjolfsson, E. and Hitt, L.M., 2000. Beyond computation: information technology, organizational
transformation and business performance, Journal of Economic Perspectives, 14 (4), 23-48.
Brynjolfsson, E. and Hitt, L.M., 2003. Computing productivity: firm-level evidence, Review of
Economics and Statistics, 85 (4), 793-808.
Brynjolfsson, E. and Kemerer, C.F., 1996. Network externalities in microcomputer software: an
econometric analysis of the spreadsheet market. Management Science, 42 (12), 1627-1647.
Calestous, J. and Lee, Y.C., 2005. Innovation: Applying Knowledge in Development. UN Millennium
Project, Task Force on Science, Technology and Innovation, Earthscan, London, UK.
Cameron, A.C. and Trivedi, P.K., 2005. Microeconometrics: Methods and Application. Cambridge
University Press, Cambridge, UK.
Chang, Y.B. and Gurbaxani, V., 2013. An empirical analysis of technical efficiency: the role of IT
intensity and competition. Information Systems Research, 24 (3), 561-578.
25
Chari, M.D.R., Devaraj, S., and David, P., 2008. The impact of information technology investments and
diversification strategies on firm performance. Management Science, 54 (1), 224-234.
Cheng, Z. and Nault, B.R., 2007. Industry level supplier-driven IT spillovers. Management Science, 53
(8), 1199-1216.
Chou, Y.C., Shao, B.B.M., and Lin, W.T., 2012. Performance evaluation of production of IT capital
goods across OECD countries: a stochastic frontier approach to Malmquist index, Decision Support
Systems, 54 (1), 173-184.
Chou, Y.C. and Shao, B.B.M., 2014. Total factor productivity growth in information technology services
industries: a multi-theoretical perspective, Decision Support Systems, 62 (1), 106-118.
Chwelos, P., Ramirez, R., Kraemer, K.L., and Melville, N.P., 2010. Does technological progress alter the
nature of information technology as a production input? New evidence and new results. Information
Systems Research, 21 (2), 392-408.
Comin, D., 2008. Total factor productivity, in S. N. Durlauf and L. E. Blume (eds), New Palgrave
Dictionary of Economics. Oakgrave MacMillian, New York, NY.
Davenport, T.H., Prusak, L., and Strong, B., 2008, March 11. Putting ideas to work: knowledge
management can make a difference—but it needs to be more pragmatic, Wall Street Journal.
Dedrick, J., Gurbaxani, V., and Kraemer, K.L., 2003. Information technology and economic performance:
a critical review of empirical evidence. ACM Computing Surveys, 35 (1), 1-28.
Delong, J.B. and Summers, L.H., 1991. Equipment investment and economic growth. Quarterly Journal
of Economics, 106 (2), 445-502.
Demeter, K., Chikán, A., and Matyusz, Z., 2011. Labour productivity change: drivers, business impact
and macroeconomic moderator. International Journal of Production Economics, 131 (1), 215-223.
Devaraj, S. and Kohli, R., 2003. Performance impacts of information technology: is actual usage the
missing link? Management Science, 49 (3), 273-289.
Dewan, S. and Min, C.K., 1997. The substitution of information technology for other factors of
production: a firm-level analysis. Management Science, 43 (12), 1660-1675.
Dewan, S. and Kraemer, K.L., 2000. Information technology and productivity: evidence from country-
level data. Management Science, 46 (4), 548-562.
Diewert, E. and Fox, K., 2001. Productivity paradox and mismeasurement of economic activity, in K.
Okina and T. Inoue (eds), Monetary Policy in a World of Knowledge-Based Growth, Quality
Change and Uncertain Measurement, Palgrave, New York, 175-197.
Freeman, C. and Perez, C., 1988. Structural crises of adjustment, business cycles and investment
behaviour, in G. Dosi, C. Freeman, R. Nelson, and L. Soete (eds), Technical Change and Economic
Theory, Pinter, London, UK.
Furman, J.L., Porter, M.E., and Stern, S., 2002. The determinants of national innovative capacity.
Research Policy, 31 (6), 899-933.
Gandal, N., 1994. Hedonic price indexes for spreadsheets and an empirical test for network externalities.
RAND Journal of Economics, 25 (1), 160-170.
Ghose, A., Smith, M.D., and Telang, R., 2006. Internet exchange for used books: an empirical analysis of
product cannibalization and welfare impact. Information Systems Research, 17 (1), 3-19.
Glass, A.J. and Saggi, K., 1998. International technology transfer and the technology gap. Journal of
Development Economics, 55 (2), 369-398.
Greene, W., 2000. Econometric Analysis, Prentice-Hall, New York, NY.
Grossman, G. and Helpman, E., 1994. Endogenous innovations in the theory of growth. Journal of
Economic Perspectives, 8 (1), 23-44.
Hajji, A. Pellerin, R., Léger, P., Gharbi, A., and Babin, G., 2012. Dynamic pricing models for ERP
26
systems under network externality. International Journal of Production Economics, 135 (2), 708-
715.
Hall, R. and Jones, C., 1999. Why do some countries produce so much more output per worker than
others? Quarterly Journal of Economics, 114 (1), 83-116.
Han, K., Chang, Y.B., and Hahn, J., 2011. Information technology spillover and productivity: the role of
information technology intensity and competition. Journal of Management Information Systems, 28
(1), 115-145.
Helpman, E., 1998. General Purpose Technologies and Economic Growth. MIT Press, Cambridge, MA.
Hoechle, D., 2007. Robust standard errors for panel regression with cross-sectional dependence. Stata
Journal, 7 (3), 281-312.
Hulten, C.R., 2001. Total factor productivity: a short biography, in C.R. Hulten, E.R. Dean, and M.J.
Harper (eds), New Developments in Productivity Analysis, University of Chicago Press, Chicago,
IL.
Jitpaiboon, T., Dobrzykowski, D.D., Ragu-Nathan, T.S., and Vonderembse, M.A., 2013. Unpacking IT
use and integration for mass customization: a service-dominant logic view. International Journal of
Production Research, 51 (8), 2527-2547.
Jorgenson, D.W., 2001. Information technology and the U.S. economy. American Economic Review, 91
(1), 1-32.
Jorgenson, D.W., 2009. Introduction, in D.W. Jorgenson (Eds.), Economics of Productivity, Edward Elgar
Publishing, UK.
Jorgenson, D.W., Ho, M., and Stiroh, K., 2006. The sources of the second surge of U.S. productivity and
implications for the future, Working Paper.
Jorgenson, D.W., Ho, M., and Stiroh, K., 2008. A retrospective look at the U.S. productivity growth
resurgence. Journal of Economic Perspectives, 22 (1), 3-24.
Joshi, K.D., Chi, L., Datta, A., and Han, S., 2010. Changing the competitive landscape: continuous
innovation through IT-leveraged knowledge capabilities. Information Systems Research., 21 (3),
472-495.
Katz, M.L. and Shapiro, C., 1985. Network externalities, competition, and compatibility. American
Economic Review, 75 (3), 424-440.
Katz, M.L. and Shapiro, C., 1986. Technology adoption in the presence of network externalities. Journal
of Political Economy, 94 (4), 822-884.
Kelly, K., 1999. New Rules for the New Economy, Penguin, New York, NY.
Kim, G., Shin, B., Kim, K.K., and Lee, H.G., 2011. IT capabilities, process-oriented dynamic capabilities,
and firm financial performance. Journal of the Association for Information Systems, 12 (7), 487-
517.
Kim, S.W. and Narasimhan, R., 2002. Information system utilization in supply chain integration efforts.
International Journal of Production Research, 40 (18), 4585–4609.
Kohli, R. and Grover, V., 2008. Business value of IT: an essay on expanding research directions to keep
up with the times. Journal of the Association for Information Systems, 9(1), 2008, 23-39.
Lin, W.T., 2009. The business value of information technology as measured by technical efficiency:
evidence from country-level data. Decision Support Systems, 46 (4), 865-874.
Lin, W.T. and Chiang, C.Y., 2011. The Impacts of country characteristics upon the value of information
technology as measured by productive efficiency. International Journal of Production Economics,
132 (1), 13-33.
Lin, W.T. and Shao, B.B.M., 2000. Relative sizes of information technology investments and productive
efficiency: their linkage and empirical evidence. Journal of the Association for Information
27
Systems, 1 (7), 1-35.
Lin, W.T. and Shao, B.B.M., 2006a. The business value of information technology and inputs substitution:
the productivity paradox revisited. Decision Support Systems, 42 (2), 493-507.
Lin, W.T. and Shao, B.B.M., 2006b. Assessing the input effect on productive efficiency in production
systems: the value of information technology capital. International Journal of Production Research,
44 (9), 1799-1819.
Lucas, H.C., Jr., 1999. Information Technology and the Productivity Paradox, Oxford University Press,
New York.
Malhotra, A., Gosain, S., and Sawy, O., 2005. Absorptive capacity configurations in supply chains:
gearing for partner-enabled market knowledge creation. MIS Quarterly, 29 (1), 145-187.
Mansfield, E., 1985. Patents and innovations: an empirical study. Management Science, 32 (2), 173-181.
Martinsons, M.G. and Martinsons, V., 2002. Rethinking the value of IT, again. Communications of the
ACM, 45 (7), 25-26.
McAfee, A. and Brynjolfsson, E., 2008. Investing in the IT that makes a competitive difference. Harvard
Business Review, 86 (7/8), 98-107.
McKinsey and Company, 2002. U.S. productivity report: 1995-2000. Retrieved from
www.mckinsey.com/knowledge.
Melville, N. and Ramirez, R., 2008. Information technology innovation diffusion: an information
requirements paradigm. Decision Support Systems, 18 (3), 247-273.
Miozzo, M. and Walsh, V., 2006. International Competitiveness and Technological Change, Oxford
University Press, New York, NY.
Nerkar, A. and Shane, S., 2007. Determinants of invention commercialization: an empirical examination
of academically sourced inventions. Strategic Management Journal, 28 (11), 1155-1166.
OECD, 2000. A New Economy? The Changing Role of Innovation and Information Technology in Growth,
Paris, France.
OECD, 2008. OECD Manual Measuring Capital, Committee on Statistics, Geneva, Switzerland.
Oliner, D.S. and Sichel, D.E., 2000. The resurgence of growth in the late 1990s: is information
technology the story? Journal of Economic Perspectives, 14 (4), 3-22.
Oren, S. and Smith, S., 1981. Critical mass and tariff structure in electronic communications markets. Bell
Journal of Economics, 12 (2), 467-486.
Park, J., Shin, S.K., and Sanders, G.L., 2007. Impact of international information technology transfer on
national productivity. Information Systems Research, 18 (1), 86-102.
Perunović, Z., Mefford, R., and Christoffersen, M., 2012. Impact of information technology on vendor
objectives, capabilities, and competences in contract electronic manufacturing. International
Journal of Production Economics, 139 (1), 207-219.
Pesaran, M., 2004. General diagnostic tests for cross section dependence in panels. Working Paper,
University of Cambridge, Cambridge, UK.
Prajogo, D. and Olhager, J., 2012. Supply chain integration and performance: the effects of long-term
relationship, information technology and sharing, and logistics integration. International Journal of
Production Economics, 135 (1), 514-522.
Rohlfs, J., 2001. Bandwagon Effects in High-Technology Industries. MIT Press, Cambridge, MA.
Romer, P.M., 1986. Increasing returns and long run growth. Journal of Political Economy 94 (5), 1002-
1037.
Romer, P.M., 1990. Endogenous technological change. Journal of Political Economy, 98 (5), 71-102.
Romer, D., 2000. Advanced Macroeconomics, 2nd
edition. McGraw-Hill, Boston, MA.
28
Schreyer, P., 2000. The contribution of information and communication technology to output growth: a
study of the G7 countries. OECD Science, Technology and Industry Working Paper, OECD
Publishing
Schreyer, P. and Webb, C., 2006. Capital stock at the OECD: status and outlook. OECD, Geneva,
Switzerland.
Sena, V., 2004. Total factor productivity and the spillover hypothesis: some new evidence. International
Journal of Production Economics, 92 (1), 31-42.
Shapiro, C. and Varian, H.R., 1999. Information Rules: A Strategic Guide to the Network Economy.
Harvard Business School Press, Boston, MA.
Shy, O., 2001. The Economics of Network Industries. Cambridge University Press, Cambridge, UK.
Sichel, D.E., 1997. A Contribution to the Theory of Economic Growth. Brookings Institution, Washington,
DC.
Solow, R., 1956. A contribution to the theory of economic growth. Quarterly Journal of Economics, 70
(1), 65-94.
Soroor, J., Tarokh, M.J., and Keshtgary, M., 2009. Preventing failure in IT-enabled systems for supply
chain management. International Journal of Production Research, 47 (23), 6543–6557.
StataCorp, 2013. Stata Statistical Software: Release 13. College Station, TX.
Stiroh, K.J., 1999. Is there a new economy? Challenge, 42 (4), 82-101.
Stiroh, K.J., 2002. Are ICT spillovers driving the new economy? Review of Income and Wealth, 48 (1),
33-57.
Triplett, J.E., 1999. The Solow productivity paradox: what do computers do to productivity? Canadian
Journal of Economics, 32 (2), 309-334.
Varian, H.R., Farrell, J., and Shapiro, C., 2004. Economics of Information Technology: An Introduction.
Cambridge University Press, New York.
WITSA, 2010. Digital Planet. World Information Technology and Services Alliance, Vienna, VA.
Wooldridge, J.M., 2001. Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge,
MA.